Advanced Lithium-Ion Battery Prognostics: A Comprehensive Review of Machine Learning Approaches for Remaining Useful Life Prediction

Year : 2025 | Volume : 03 | Issue : 02 | Page : 12 27
    By

    Kamlesh Sharadchandra Mahajan,

  • Nitish Kumar Gautam,

  1. PhD Research Scholar, Department of Mechanical Engineering, Shri Jagdishprasad Jhabarmal Tibrewala University, Vidyanagri, Jhunjhunu 333010, Rajasthan, India
  2. Assistant Professor and PhD Research Guide, Department of Mechanical Engineering, Shri Jagdishprasad Jhabarmal Tibrewala University, Vidyanagri, Jhunjhunu 333010, Rajasthan, India

Abstract

The lithium-ion battery (LIB), as one of the main sources for portable power systems, has been increasingly popular owing to its widespread applications in electric vehicles, consumer electronics, aerospace and renewable energy. Despite their advantages in high energy density and long cycle life, LIBs suffer from degradation over time of aging and cycling, resulting in loss of performance, safety issues, and economic bottlenecks. Predicting their Remaining Useful Life (RUL) is critical for proactive maintenance, cost savings, and prolonging the operational life span. Classical physics-based models face a challenge to capture the complex nonlinear degradation patterns of LIBs, which stem from multiple coupled electrochemical processes like SEI formation, lithium plating, and mechanical stress and electrolyte decomposition. These are compounded by external influences such as temperature, depth of discharge and charge rates. Machine Learning (ML) provides an attractive alternative by learning degradation mode from data without using explicit physical equations. Models such as SVR, Random Forest (RF), Gaussian Process Regression (GPR) and deep learning models like LSTM, CNN, hybrid networks have achieved great success in estimating RUL. The emerging trends are Transfer Learning, Survival Analysis and Explainable AI (XAI) for generalization, reliability and interpretability of the models under a wide diversity of chemistries and real-world conditions. This is all done despite outstanding difficulties such as modelling the capacity recovery, uncertainty quantification and varying cell form factors. Such frameworks as Battery ML provide a glimpse of how domain-aware engineering is already beginning to integrate with AI, evolving toward intelligent lifecycle management of next-gen battery systems.

Keywords: Lithium-ion Batteries (LIBs), Remaining Useful Life (RUL) Prediction, Machine Learning (ML), Battery Degradation Mechanisms, Explainable Artificial Intelligence (XAI)

[This article belongs to International Journal of Mechanical Dynamics and Systems Analysis ]

How to cite this article:
Kamlesh Sharadchandra Mahajan, Nitish Kumar Gautam. Advanced Lithium-Ion Battery Prognostics: A Comprehensive Review of Machine Learning Approaches for Remaining Useful Life Prediction. International Journal of Mechanical Dynamics and Systems Analysis. 2025; 03(02):12-27.
How to cite this URL:
Kamlesh Sharadchandra Mahajan, Nitish Kumar Gautam. Advanced Lithium-Ion Battery Prognostics: A Comprehensive Review of Machine Learning Approaches for Remaining Useful Life Prediction. International Journal of Mechanical Dynamics and Systems Analysis. 2025; 03(02):12-27. Available from: https://journals.stmjournals.com/ijmdsa/article=2025/view=234144


References

[1] Z. Zhang, Y. Liu, and H. Wang, “Battery health monitoring and remaining useful life prediction techniques: A review of technologies,” MDPI, [Online]. Available: https://www.mdpi.com .

[2] BatteryML Consortium, “BatteryML: An open-source platform for machine learning on battery degradation,” arXiv, no. 2405.00055, 2024. [Online]. Available: https://arxiv.org/abs/2405.00055 .

[3] BatteryML Consortium, “BatteryML: An open-source platform for machine learning on battery degradation,” arXiv, no. 2405.00055, 2024. [Online]. Available: https://arxiv.org/abs/2405.00055 .

[4] X. Li et al., “Battery state of health estimation under fast charging via deep transfer learning,” PMC – PubMed Central, 2023. [Online]. Available: https://www.ncbi.nlm.nih.gov/pmc .

[5] A. Kumar and S. Singh, “Comparison of developed ML models for predicting the RUL of Li-ion batteries,” ResearchGate. [Online]. Available: https://www.researchgate.net .

[6] T. Nguyen and R. Chen, “DLinear-based prediction of remaining useful life of lithium-ion batteries: Feature engineering through explainable artificial intelligence,” arXiv , no. 2405.16889, 2024. [Online]. Available: https://arxiv.org/abs/2405.16889 .

[7] T. Nguyen and R. Chen, “DLinear-based prediction of remaining useful life of lithium-ion batteries: Feature engineering through explainable artificial intelligence,” ResearchGate, 2024. [Online]. Available: https://www.researchgate.net/publication/380386134 .

[8] J. Kim et al., “DiffBatt: A diffusion model for battery degradation prediction and synthesis,” arXiv , no. 2406.07355, 2024. [Online]. Available: https://arxiv.org/abs/2406.07355 .

[9] F. Müller, “Early-cycle internal impedance enables ML-based battery cycle life predictions across manufacturers,” arXiv, no. 2401.05656, 2024. [Online]. Available: https://arxiv.org/abs/2401.05656 .

[10] M. Wei et al., “Early-stage state-of-health prediction of lithium batteries for wireless sensor networks using LSTM and a single exponential degradation model,” PMC – PubMed Central, 2023. [Online]. Available: https://www.ncbi.nlm.nih.gov/pmc .

[11] Monolith AI, “Evolution of battery models: Electrochemical to data-driven,” [Online]. Available: https://monolithai.com .

[12] S. Ali et al., “Explainable artificial intelligence (XAI) in battery management systems: A comprehensive review,” Scientific Research Archives, 2023. [Online]. Available: https://sciresjournals.com .

[13] H. Zhang et al., “HybridoNet-Adapt: A domain-adapted framework for accurate lithium-ion battery RUL prediction,” arXiv , no. 2503.21392, 2025. [Online]. Available: https://arxiv.org/abs/2503.21392 .

[14] G. Liu et al., “A data-driven approach with error compensation and uncertainty quantification for remaining useful life prediction of lithium-ion battery,” SciSpace , 2022. [Online]. Available: https://www.scispace.com .

[15] Y. Tan et al., “A data-driven framework for lithium-ion battery RUL using LSTM and XGBoost with feature selection via binary firefly algorithm,” IDEAS/RePEc, 2024. [Online]. Available: https://ideas.repec.org .

[16] R. Patel et al., “A hybrid approach for lithium-ion battery remaining useful life prediction using signal decomposition and machine learning,” PMC – PubMed Central, 2023. [Online]. Available: https://www.ncbi.nlm.nih.gov/pmc .

[17] D. Kim et al., “A machine learning tool to investigate lithium-ion battery degradation under real automotive conditions,” ResearchGate, 2023. [Online]. Available: https://www.researchgate.net .

[18] A. Khan et al., “A novel approach for enhancing battery reliability in market using machine learning-based RUL prediction,” ResearchGate, 2023. [Online]. Available: https://www.researchgate.net .

[19] L. Chen et al., “A review of lithium-ion battery state of charge estimation methods,” MDPI. [Online]. Available: https://www.mdpi.com .

[20] P. Gupta and R. Sharma, “A review of modern machine learning techniques in the prediction of remaining useful life of lithium-ion batteries,” MDPI, 2023. [Online]. Available: https://www.mdpi.com .

[21] P. Gupta and R. Sharma, “A review of modern machine learning techniques in the prediction of remaining useful life of lithium-ion batteries,” ResearchGate, 2023. [Online]. Available: https://www.researchgate.net/publication/366875874 .

[22] International Institute for Applied Systems Analysis, “A comparative study of different battery geometries used in electric vehicles,” IIASA PURE, Technical Report RR-23-004, [Online]. Available: https://pure.iiasa.ac.at .

[23] M. Hussain et al., “A critical review of online battery remaining useful lifetime prediction methods,” Frontiers, 2023. [Online]. Available: https://www.frontiersin.org .

[24] K. Lee et al., “ACCEPT: Diagnostic forecasting of battery degradation through contrastive learning,” arXiv , no. 2501.05656, 2025. [Online]. Available: https://arxiv.org/abs/2501.05656 .

[25] R. Thomas et al., “Key considerations for cell selection in electric vertical takeoff and landing vehicles: A perspective,” RSC Publishing, 2024. [Online]. Available: https://pubs.rsc.org/en/content/articlelanding/2024/ee/d4eb00024b .

[26] Monolith AI, “Optimising battery aging studies with ML,” [Online]. Available: https://monolithai.com .

[27] H. Zhang et al., “Optimizing lithium-ion battery discharge capacity prediction using Light GBM and explainable AI (XAI) framework,” Metallurgical and Materials Engineering, 2023. [Online]. Available: https://metall-mater-eng.com .

[28] Y. Zhao et al., “Physics-informed neural network for lithium-ion battery degradation stable modeling and prognosis,” PMC – PubMed Central, 2023. [Online]. Available: https://www.ncbi.nlm.nih.gov/pmc .

[29] C. Park et al., “Predicting the RUL of Li-ion batteries in UAVs using machine learning techniques,” MDPI, [online]. Available: https://www.mdpi.com .

[30] Grepow Battery Company, “Prismatic vs pouch vs cylindrical lithium ion battery cell,” [Online]. Available: https://grepow.com .

[31] W. Xu et al., “Remaining useful life prediction of lithium-ion batteries using neural networks with adaptive Bayesian learning,” PMC – PubMed Central, 2023. [Online]. Available: https://www.ncbi.nlm.nih.gov/pmc .

[32] M. Rahman et al., “Remaining useful life prediction using machine learning models: A comparative study,” ResearchGate, 2022. [Online]. Available: https://www.researchgate.net/publication/361619675 .

[33] M. Rahman et al., “Remaining useful life prediction using machine learning models: A comparative study,” MDPI, 2022. [Online]. Available: https://www.mdpi.com .

[34] A. Hassan et al., “Remaining useful life prediction with uncertainty quantification using evidential deep learning,” ResearchGate, 2023. [Online]. Available: https://www.researchgate.net/publication/369475086 .

[35] S. Das et al., “Remaining useful life prediction for batteries utilizing an explainable AI approach with a predictive application for decision-making,” arXiv , no. 2405.01083, 2024. [Online]. Available: https://arxiv.org/abs/2405.01083 .

[36] B. Kim et al., “RUL prediction for lithium battery systems in fuel cell ships based on adaptive modal enhancement networks,” MDPI, 2023. [Online]. Available: https://www.mdpi.com .

[37] J. Lee et al., “Survival analysis with machine learning for predicting Li-ion battery remaining useful life,” arXiv , no. 2501.05656, 2025. [Online]. Available: https://arxiv.org/abs/2501.05656 .

[38] T. Walker et al., “The importance of degradation mode analysis in parameterising lifetime prediction models of lithium-ion battery degradation,” PMC – PubMed Central, 2023. [Online]. Available: https://www.ncbi.nlm.nih.gov/pmc .

[39] S. Ali et al., “Transfer learning-based hybrid remaining useful life prediction for lithium-ion batteries under different stresses,” ResearchGate, 2023. [Online]. Available: https://www.researchgate.net .

[40] Newswise, “Unlocking the secrets of battery degradation: A transfer learning approach,” [Online]. Available: https://www.newswise.com .

[41] BatteryML Consortium, “End-to-end framework for predicting the remaining useful life of lithium-ion batteries,” arXiv , no. 2505.16664, 2025. [Online]. Available: https://arxiv.org/abs/2505.16664 .

[42] GreyB Services, “Pressure monitoring of EV battery cells,” [Online]. Available: https://xray.greyb.com .

[43] H. Zhou et al., “An interpretable online prediction method for remaining useful life of lithium-ion batteries,” PMC – PubMed Central, 2023. [Online]. Available: https://www.ncbi.nlm.nih.gov/pmc .

Regular Issue Subscription Original Research
Volume 03
Issue 02
Received 27/10/2025
Accepted 03/12/2025
Published 15/12/2025
Publication Time 49 Days
6

Login


My IP

PlumX Metrics